In this demo paper, we introduce the Multimodal Pipeline, a prototypical approach for the collection, storing, annotation, processing and exploitation of multimodal data for supporting learning. At the current stage of development, the Multimodal Pipeline consists of two relevant prototypes: 1) Multimodal Learning Hub for the collection and storing of sensor data from multiple applications and 2) the Visual Inspection Tool for visualisation and annotation of the recorded sessions. The Multimodal Pipeline is designed to be a flexible system useful for supporting psychomotor skills in a variety of learning scenarios such as presentation skills, medical simulation with patient manikins or calligraphy learning. The Multimodal Pipeline can be configured to serve different support strategies, including detecting mistakes and prompting live feedback in an intelligent tutoring system or stimulating self-reflection through a learning analytics dashboard.
|Number of pages||3|
|Publication status||Published - 10 Aug 2019|
|Event||First workshop on AI-based Multimodal Analytics for Understanding Human Learning in Real-world Educational Contexts - Macau, China|
Duration: 11 Jan 2020 → 11 Jan 2020
Conference number: 1
|Workshop||First workshop on AI-based Multimodal Analytics for Understanding Human Learning in Real-world Educational Contexts|
|Period||11/01/20 → 11/01/20|
Di Mitri, D., Schneider, J., Specht, M. M., & Drachsler, H. J. (2019). Multimodal Pipeline: A generic approach for handling multimodal data for supporting learning. Paper presented at First workshop on AI-based Multimodal Analytics for Understanding Human Learning in Real-world Educational Contexts, Macau, China.